Mega8535 & Mega16 dead once Mega164 available?

Go To Last Post
2 posts / 0 new
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DigiKey now has posted prices (finally) for Mega164 and Mega324. (Stock isn't expected for another two months, but I've gotten samples.)

Let's take a peek at posted DigiKey prices in qty. 100, US$:

ATmega164 $3.18
ATmega324 $3.98
ATmega644V $4.87
ATmega644P $4.51 (a couple of models)
ATmega644PV $5.13

ATmega32 $4.75
ATmega16 $3.75
ATmega8535 $3.25

As you can see, yet another anomaly in the AVR pricing structure (as Mega329 cheaper than Mega169).

We dropped a Mega644V into a Mega16L design without any problems, so I am not anticipating major upheaval when using Mega164 instead of a Mega16 or '8535.

Are there any gotchas that are not readily apparent? For example, do the smaller die sizes of the new generation that lead to better yields and better prices also mean less noise immunity?

On the surface, the new generation:
-- less expensive
-- "P" features
-- 20MHz max clock rate
-- twin USARTs, and USARTs as SPI masters
-- pin-change interrupts
-- "V" features (e.g. watchdog interrupt; low-power modes)
-- [vs. '8535] JTAG; ADC gain

These look like all advantages for the new generation. Are there any disavantages that I missed?


You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Some more differences:

An output pin state can be toggled by writing to the pin register on the new chips.

Added new GPIOR0, GPIOR1 and GPIOR2 registers.

Timer/counter 0 now has a second output compare.

The new asynchronous timer/counter 2 can use an external oscillator instead of only a watch crystal.

The only somewhat selective disadvantage I can see is related to the numerous changes in the new register organization that will impact porting assembly programs much more than compiled programs. The new chips have registers that have been moved outside the CBI/SBI address range (many beyond the IN/OUT address range). I really can't see a situation where an existing algorithm is so optimized that the extra register access instruction cycles will cause any problem.

I also noticed DigiKey added pricing for the ATtiny24/44/84, ATmega640 and AT90CAN32/64. It looks like the new U green version of the AT90CAN128 will take a big price drop.

BTW, happy holidays to everyone that celebrates this time of year.